Learning Dynamic Graph Representations through Timespan View Contrasts

2026-05-26

AI Summary

Existing static graph representation methods struggle to model the evolution of dynamic graphs due to their neglect of temporal information. This work proposes a dynamic graph representation learning framework based on temporal interval view contrast, incorporating time-translation invariance as an inductive bias. The authors introduce CLDG and its enhanced variant CLDG++, which integrate graph diffusion mechanisms with multi-scale local-global contrastive learning to implicitly capture temporal evolution cues. The proposed approach achieves state-of-the-art performance on node classification and dynamic graph anomaly detection tasks while significantly reducing both time and space complexity.

Abstract

The rich information underlying graphs has inspired further investigation of unsupervised graph representation. Existing studies mainly depend on node features and topological properties within static graphs to create self-supervised signals, neglecting the temporal components carried by real-world graph data, such as timestamps of edges. To overcome this limitation, this paper explores how to model temporal evolution on dynamic graphs elegantly. Specifically, we introduce a new inductive bias, namely temporal translation invariance, which illustrates the tendency of the identical node to keep similar labels across different timespans. Based on this assumption, we develop a dynamic graph representation framework CLDG that encourages the node to maintain locally consistent temporal translation invariance through contrastive learning on different timespans. Except for standard CLDG which only considers explicit topological links, our further proposed CLDG++ additionally employs graph diffusion to uncover global contextual correlations between nodes, and designs a multi-scale contrastive learning objective composed of local-local, local-global, and global-global contrasts to enhance representation capabilities. Interestingly, by measuring the consistency between different timespans to shape anomaly indicators, CLDG and CLDG++ are seamlessly integrated with the task of spotting anomalies on dynamic graphs, which has broad applications in many high-impact domains, such as finance, cybersecurity, and healthcare. Experiments demonstrate that CLDG and CLDG++ both exhibit desirable performance in downstream tasks including node classification and dynamic graph anomaly detection. Moreover, CLDG significantly reduces time and space complexity by implicitly exploiting temporal cues instead of complicated sequence models.

Problem

• dynamic graphs
• temporal evolution
• graph representation learning
• time-aware modeling
• anomaly detection

Innovation

• temporal translation invariance
• contrastive learning
• dynamic graph representation
• graph diffusion
• multi-scale contrast